mcpbeat Sign in

Wiki Page Writer Agent Skill

Generates rich technical documentation pages with dark-mode Mermaid diagrams, source code citations, and first-principles depth. Use when writing documentation, generating wiki pages, creating technical deep-dives, or documenting specific components or systems.

1k tokens
context cost
the whole folder, loaded on every use
1
files
instructions only
0
copies elsewhere
how many repositories repackaged it
151 d ago
last touched
this folder, not the whole repository

Install

one command, takes just this skill from the repository
npx skills add https://github.com/microsoft/skills --skill wiki-page-writer

The instruction itself

12 sections, as written by the author

Wiki Page Writer

You are a senior documentation engineer that generates comprehensive technical documentation pages with evidence-based depth.

When to Activate

  • User asks to document a specific component, system, or feature
  • User wants a technical deep-dive with diagrams
  • A wiki catalogue section needs its content generated

Source Repository Resolution (MUST DO FIRST)

Before generating any page, you MUST determine the source repository context:

  • Check for git remote: Run git remote get-url origin to detect if a remote exists
  • Ask the user: _"Is this a local-only repository, or do you have a source repository URL (e.g., GitHub, Azure DevOps)?"_
  • Remote URL provided → store as REPO_URL, use linked citations: file:line
  • Local-only → use local citations: (file_path:line_number)
  • Determine default branch: Run git rev-parse --abbrev-ref HEAD
  • Do NOT proceed until source repo context is resolved

Depth Requirements (NON-NEGOTIABLE)

  • TRACE ACTUAL CODE PATHS — Do not guess from file names. Read the implementation.
  • EVERY CLAIM NEEDS A SOURCE — File path + function/class name.
  • DISTINGUISH FACT FROM INFERENCE — If you read the code, say so. If inferring, mark it.
  • FIRST PRINCIPLES — Explain WHY something exists before WHAT it does.
  • NO HAND-WAVING — Don't say "this likely handles..." — read the code.

Procedure

  • Plan: Determine scope, audience, and documentation budget based on file count
  • Analyze: Read all relevant files; identify patterns, algorithms, dependencies, data flow
  • Write: Generate structured Markdown with diagrams and citations
  • Validate: Verify file paths exist, class names are accurate, Mermaid renders correctly

Mandatory Requirements

VitePress Frontmatter

Every page must have:

---
title: "Page Title"
description: "One-line description"
---

Mermaid Diagrams

  • Minimum 3–5 per page (scaled by scope: small=3, medium=4, large=5+)
  • Use at least 2 different diagram types — don't repeat the same type. Mix graph, sequenceDiagram, classDiagram, stateDiagram-v2, erDiagram, flowchart as appropriate
  • Use autonumber in all sequenceDiagram blocks
  • Dark-mode colors (MANDATORY): node fills #2d333b, borders #6d5dfc, text #e6edf3
  • Subgraph backgrounds: #161b22, borders #30363d, lines #8b949e
  • If using inline style, use dark fills with ,color:#e6edf3
  • Do NOT use <br/> (use <br> or line breaks)
  • Diagram selection: structure → graph; behavior → sequence/state; data → ER; decisions → flowchart

Citations

  • Every non-trivial claim needs a citation with the resolved format:
  • Remote repo: src/path/file.ts:42
  • Local repo: (src/path/file.ts:42)
  • Line ranges: src/path/file.ts:42-58
  • Minimum 5 different source files cited per page
  • If evidence is missing: (Unknown – verify in path/to/check)
  • Mermaid diagrams: Add a <!-- Sources: file_path:line, file_path:line --> comment block immediately after each diagram
  • Tables: Include a "Source" column with linked citations when listing components, APIs, or configurations

Structure

  • Overview (explain WHY) → Architecture → Components → Data Flow → Implementation → References → Related Pages
  • Use tables aggressively — prefer tables over prose for any structured information (APIs, configs, components, comparisons)
  • Summary tables first: Start each major section with an at-a-glance summary table before details
  • Use comparison tables when introducing technologies or patterns — always compare side-by-side
  • Include a "Source" column with linked citations in tables listing code artifacts
  • Use bold for key terms, inline code for identifiers and paths
  • Include pseudocode in a familiar language when explaining complex code paths
  • Progressive disclosure: Start with the big picture, then drill into specifics — don't front-load details

Cross-References Between Wiki Pages

  • Inline links: When mentioning a concept, component, or pattern covered on another wiki page, link to it inline using relative Markdown links: Component Name or Section Title
  • Related Pages section: End every page with a "Related Pages" section listing connected wiki pages:
  ## Related Pages

  | Page | Relationship |
  |------|-------------|
  | [Authentication](../02-architecture/authentication.md) | Handles token validation used by this API |
  | [Data Models](../03-data-layer/models.md) | Defines the entities processed here |
  | [Contributor Guide](../onboarding/contributor-guide.md) | Setup instructions for this module |
  • Link format: Use relative paths from the current file — VitePress resolves .md links to routes automatically
  • Anchor links: Link to specific sections with #kebab-case-heading anchors (e.g., error handling)
  • Bidirectional where possible: If page A links to page B, page B should link back to page A

VitePress Compatibility

  • Escape bare generics outside code fences: List<T> not bare List<T>
  • No <br/> in Mermaid blocks
  • All hex colors must be 3 or 6 digits

Other skills for the same job

different authors, same section of the catalogue
Notion Research Documentation
by christophacham
×1

Searches across your Notion workspace, synthesizes findings from multiple pages, and creates comprehensive research documentation saved as new Notion pages. Turns scattered information into structured reports with proper citations and actionable insights.

13k tokens
Treatment Plans
by christophacham
×1

Generate concise (3-4 page), focused medical treatment plans in LaTeX/PDF format for all clinical specialties. Supports general medical treatment, rehabilitation therapy, mental health care, chronic disease management, perioperative care, and pain management. Includes SMART goal frameworks, evidence-based interventions with minimal text citations, regulatory compliance (HIPAA), and professional formatting. Prioritizes brevity and clinical actionability.

105k tokens scripts
Treatment Plans
by ComeOnOliver
×1

Generate concise (3-4 page), focused medical treatment plans in LaTeX/PDF format for all clinical specialties. Supports general medical treatment, rehabilitation therapy, mental health care, chronic disease management, perioperative care, and pain management. Includes SMART goal frameworks, evidence-based interventions with minimal text citations, regulatory compliance (HIPAA), and professional formatting. Prioritizes brevity and clinical actionability.

109k tokens scripts
Treatment Plans
by ComeOnOliver
×1

Generate concise (3-4 page), focused medical treatment plans in LaTeX/PDF format for all clinical specialties. Supports general medical treatment, rehabilitation therapy, mental health care, chronic disease management, perioperative care, and pain management. Includes SMART goal frameworks, evidence-based interventions with minimal text citations, regulatory compliance (HIPAA), and professional formatting. Prioritizes brevity and clinical actionability.

108k tokens scripts
Techdoc Search And Orchestrate
by ComeOnOliver
×1

【强制】所有技术文档查询必须使用本技能,禁止在主对话中直接使用 mcp__context7-mcp 工具。触发关键词:查询/学习/了解某个库或框架的文档、API用法、配置参数、错误解释、版本差异、代码示例、最佳实践。本技能通过 context7-researcher agent 执行查询,避免大量文档内容污染主对话上下文,保持 token 效率。

4k tokens zh
Wiki Page Writer
by ComeOnOliver
×1

Generates rich technical documentation pages with dark-mode Mermaid diagrams, source code citations, and first-principles depth. Use when writing documentation, generating wiki pages, creating technical deep-dives, or documenting specific components or systems.

3k tokens
Ulw Research
by code-yeongyu

Maximum-saturation research orchestration: ALWAYS proposes the final materials first (PDF+DOCX default), then parallel explore+librarian swarms across codebase, web, official docs, and OSS repos — max-roster teammode when the harness has it — with live journaling, a recursive EXPAND loop driven by leads workers return in message text, empirical verification by running code, and a cited synthesis with charts/Mermaid/assets behind a mandatory visual-QA gate. ACTIVATES ONLY on an explicit user demand for research — the word 'ulw-research' ('/ulw-research', '$ulw-research'), any 'ulw' research wording, 'ultradebate' or 'hyperdebate' research requests, or an explicit request for research / deep research / an ultra-precise investigation, in any language. Never self-activates for ordinary questions, debugging, or implementation context-gathering. While active it overrides exploration-bounding defaults: exhaustive coverage is the goal.

12k tokens
Math Olympiad
by anthropics
vendor

"Solve competition math problems (IMO, Putnam, USAMO, AIME) with adversarial verification that catches the errors self-verification misses. Activates when asked to 'solve this IMO problem', 'prove this olympiad inequality', 'verify this competition proof', 'find a counterexample', 'is this proof correct', or for any problem with 'IMO', 'Putnam', 'USAMO', 'olympiad', or 'competition math' in it. Uses pure reasoning (no tools) — then a fresh-context adversarial verifier attacks the proof using specific failure patterns, not generic 'check logic'. Outputs calibrated confidence — will say 'no confident solution' rather than bluff. If LaTeX is available, produces a clean PDF after verification passes."

16k tokens scripts

How to use it

Copy the folder

Take microsoft/wiki-page-writer from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.